412 PROCEEDINGS OP THE AMERICAN ACADEMY. 



The various isomeric forms of Formula I. need not be considered, 

 as the arguments which follow will apply as well to them as to this 

 one. From the properties of the substance, it is nut possible to decide 

 between these formulas. It is white, like Nef's bromine addition 

 products, but an acetal like Formula II. would also probably be white,^ 

 as the colors of chloranil and its derivatives undoubtedly depend on 

 the fact that they are quinoiies, and in such an acetal as this the 

 quinone nature is obliterated. Its most striking property is the ease 

 with which the free substance loses the two additional molecules of 

 methyl alcohol, since it passes into dichlordiniethoxyquinone when 

 heated to temperatures from 160° to 195° ; the same change is brought 

 about by dilute sulphuric or hydrochloric acid, slowly in the cold. 

 more rapidly when the mixture is warmed gently. While this insta- 

 bility is in harmony with the second formula given above, it is 

 hard to connect it with the first, as it seems improbable that a methyl 

 group attached directly to the benzol ring, as in Formula I., could be 

 so easily removed. It must be remembered in this connection, how- 

 ever, that Stieglitz * by treating the dichlorquinonedimalonic ester 

 with dilute sodic hydrate in the cold obtained parachlorhydroxy- 

 quinone, CgC10HH202. As in this case carbon atoms attached to the 

 quinone ring were removed by a dilute alkali in a few minutes, it 

 might be that the methyl group in Formula I. could be removed under 

 the conditions observed by us. The objection to Formula I. mentioned 

 above might also be met by substituting for it one in which the 

 radicals added were Na and CHoO instead of NaO and CH., as in that 

 formula, but such an addition seems to us at the best improbable. It 

 should be remarked here that Nef's bromine addition compounds were 

 decomposed with the utmost ease by alkalies, but were very stable in 

 contact with acids, being crystallized from strong nitric acid ; our sub- 

 stances show just the opposite behavior, as they are rather remarkably 

 stable toward alkalies ; but this difference might be due to the differ- 

 ence in the nature of the radicals added (Br„ in one case, NaO and 

 CHg in the other) rather than to a difference in the structure of the 

 compounds. 



Since, as explained above, the properties of our new substance 

 were not sufficient to prove the correctness of one or the other of the 

 two formulas proposed, we next turned our attention to some experi- 

 ments with derivatives of this substance, or rather of the correspond- 

 ing ethyl compound, which have settled the question conclusively. 



* Am. Cliem. Journ., XIII. 38. 



